Adaptive Filter Structures (Transversal structure (Output is the linear…
Adaptive Filter Structures
Output is the linear combination of the delayed samples of the input sequence.
Single input and output with desired signal.
Commonly used structure is transversal structure.
Transversal structure refers finite-impulse response.
Infinite impulse response includes recursive equation.
Linear combiner structure
In beam forcing applications the inputs are not the delayed samples of a signal input.
The output is the linear combination of the different signals received at its tap inputs.
This structure is more general than Transversal structure.
Transversal and linear combiner structure are non-recursive filters and the output does not involve any feedback mechanism.
The lattice structures are more complicated than the direct implementations, but they have some advantages in certain applications.
FIR and IIR may alternatively be implemented using the lattice structures.
E.g, Linear prediction for speech processing where we need to realize all-pole (IIR) filters.
The lattice structure can be more easily controlled to prevent possible instability of the filter
In least mean square method, use of lattice structure leads to a computationally efficient algorithm known as recursive least squares lattice.
IIR filters have been used in many applications, but their application in the area of adaptive filters is rather limited.
They can easily become unstable because their poles may get shifted out of the unit circle by the adaptation process.
The performance function (MSE) has many local minima points and it may result in convergence of the filter to one of the local minima and not to the desired goal.
The MSE functions of FIR filter and linear combiner are well-behaved quadratic functions with a single minimum point.
Because of this feature, non recursive filters are used in most of the applications of adaptive filter.